As predictable as the sunrise in the morning, I speak every day with sales and marketing leaders who worry they’re not doing enough with AI and have fallen too far behind.
They feel like they’ve failed and are afraid they won’t live up to their leaders’ expectations. Take a deep breath. You’re not a freak failure and that doesn’t mean your team sucks.
I just read an article the other day that talked about the “early days” of AI. The author was referring to last fall and winter. While it may seem strange, it’s becoming increasingly clear how these tools are supposed to work with you and your team.
Enough of the “current state”
According to McKinsey, about 87% of companies are in the early stages of AI adoption or have not yet started. Gartner reports that only 24% of marketers say AI and machine learning are a top priority in their tools and tech stack, highlighting a significant gap in how much AI is prioritized in their budgets.
Is AI a silver bullet? Do you have highly skilled AI experts on your team? Can you afford to outsource an eight-figure project to one of the international consulting firms? Probably not.
Do you need all of these things today? Probably not. However, they can have a transformative effect on those who invest in a path forward.
What does a successful “future state” look like?
The Plan: “By failing to prepare, you are preparing to fail.”
Benjamin Franklin was right about this almost 300 years ago.
Do you have a cross-functional team to assess use cases and create policy? AI is no different than any other part of an organization; you need to be prepared and have a plan. Without a plan, you’ll end up with Bring Your Own AI (BYOAI) and chaos. AI is already happening.
We recommend that clients create a stakeholder panel with diverse interests. While you need to include the voices of your technical, HR, and legal teams, you need to surround these protectionist voices with stakeholders interested in creating value, such as sales, marketing, customer success, sales enablement, product, and so on.
Interests related to data privacy, responsible use, and careful selection of use cases are important and should be defended. However, these concerns should not be the only concern, as they may not fully understand your goals and end up hindering your progress. They need your knowledge and training.
Dig deeper: A User-Friendly Approach to Adopting AI in Marketing
The need: identify practical use cases for your team
Narrow your focus and avoid confusion and consternation. You’re not trying to set this decision in stone, but take a moment to brainstorm applications that your team will find useful. In this decision process, consider selecting an area where you have low perceived risk and high perceived value.
Many marketers choose content creation and customization for this very reason. It’s hard to get right, and the result is that one would hope that this information becomes public knowledge: IT, privacy, and legal teams wouldn’t (should) have any concerns about this and other use cases where information culture is harmless.
The goal: efficiency
When selecting use cases, think about how AI can help your team become more efficient. Generative AI is amazing when it learns hundreds of dimensions of analysis at once. The average very intelligent person can analyze two to three dimensions. That is its core strength: efficiency. AI inherently creates efficiency, so the desired future state is to create productivity (effectiveness + efficiency).
How will you identify the productivity gained for each use case? What measures will focus on effectiveness? Separately, what will measure efficiency? When the “experts” focus on efficiency, they are wrong. Unless you solve the problem efficiency Part of the equation: efficiency simply doesn’t matter.
Think about it this way: How many times have you read an email filled with AI-generated words that you intuitively know were created faster than a human could type, but were completely irrelevant? Hundreds? Yes, me too. Effective? Yes. Effective? No. And doing more of them is negative and corrosive, not positive and helpful.
Dig deeper: From efficiency to effectiveness: the B2B marketing revolution in 2024
The challenge: the context
However large language models Be aware that they are virtually useless for sales and marketing use cases. While they may tell you about quantum physics or best agricultural practices in sub-Saharan Africa, they completely lack the context of your organization, why your solution approach is better, why your capabilities are superior, or why your audience’s needs may differ from those of your competitors.
The most expensive solution is to rely on rapid engineering, downloading a few pages of content, or creating many microscopic custom GPTs. This approach is the current state of affairs and we see confusion, low adoption, and lack of impact.
Without this context, your interactions are generic, and that’s where the AI jargon begins.
Focus your solution choice on the ability to train the AI tool to your business strategy and perspective. Use rapid engineering to achieve this.
The Path: Native, Embedded, or Custom-Built AI Tools
Non-trivially, even if you don’t do anything with AI, that doesn’t mean you’re not using it. Your phone and video call transcription services all use AI.
But should adoption be guided or left to happen randomly? BYOAI is a reality, just as BYOD (Bring Your Own Device) was a decade or two ago. I typically coach organizations to see three classes of AI tools:
Native. While most people immediately think of chat-based tools like ChatGPT, there are also enterprise tools like OpenAI. Many “experts” get this wrong. You can create an OpenAI account and access OpenAI’s completely unbridled environment.
Data privacy and hallucinations are largely the result of consumer-facing chat tools, not their enterprise counterparts. Trying to make the consumer chatbot work in a business environment is problematic. It is a huge (costly) waste of time and provides only nominal value because it lacks the context of your business that can drive value.
Direct access to enterprise tools may be more expensive and require new skills, but it is THE way to experience AI.
IntegratedThink about the tools you use every day in marketing and sales: your CRM, marketing automation platform, or sales enablement tools. Embedded AI solutions fit seamlessly into these familiar environments, giving you features like predictive analytics and personalized content recommendations right at your fingertips.
But let’s be honest. While these solutions simplify repetitive tasks and provide valuable insights, they often lack the unique context of your organization’s specific needs and goals. This shortcoming limits their ability to generate real efficiency.
The key question is: Do these integrated AI solutions improve the effectiveness of your AI use cases? For example, do they make your marketing campaigns more targeted and effective, or do they simply speed up the process without adding substantial value?
Before you fully embrace these tools, evaluate how they align with your strategic goals. Do they meet the nuanced demands of your marketing and sales efforts, or do they fall short? This self-assessment is critical to determining whether these integrated options are truly generating transformative results.
Specially designedThese are bespoke solutions, usually focused on a specific use case. Because they are tailored to specific use cases, they are more likely to provide relevant context and actionable insights. But here’s the question: how well can they adapt to your unique environment?
The key here is to ensure that the context you provide is sufficient to drive meaningful effectiveness. Can you train the tools to deeply understand your capabilities and solutions, your audience, your differentiation, and your market approach?
Dig deeper: It’s time to teach your brand AI
Contributing authors are invited to create content for MarTech and are chosen for their expertise and contribution to the MarTech community. Our contributors work under the supervision of Writing and contributions are checked for quality and relevance to our readers. Opinions expressed are their own.
